Causing and "causing"
Brian argues that our intuitions about whether an action C causes somebody's continued survival is linked to the applicability of causative notions like "opening", "closing", "protecting", "threatening": if C inadvertently causes the survivor to be threatened but at the same time protects him from the threat, we are more inclined to count C as causing the survival than if C threatens the surviver but at the same time inadvertently causes him to be protected.
This has interesting consequences for the analysis of causation. One aspect of this is that whether a causative applies seems to depend on the conversational background.
My favourite example is the retina: when light hits a photoreceptor, it doesn't cause the receptor to send excitatory signals to the brain; rather, it prevents the receptor from sending inhibitory signals to bipolar cells, which in the absence of these signals auto-generate electrochemical activity that activates retinal ganglion cells connected to the cortex. (This is only very roughly what happens, but suppose for the sake of the example that it is exactly how things turned out.) Now does the incoming light cause the activity in the ganglion cells and brain? Does the light activate the ganglion cells? In some contexts, it is perfectly acceptable (and true) to say yes. Otherwise this would be a serious problem for causal theories of perception, knowledge, belief, etc. But in other contexts, it is equally acceptable (and true) to say that the light doesn't really do any causing and activating: what activates the ganglion cells is the auto-generated activity in the bipolar cells, and the light doesn't contribute to that.
Or take Brian's example of causing a door to slam shut by opening a window. If I do this without intending to close the door, we normally wouldn't say that I closed the door. I only caused it to close. On the other hand, suppose there is a law against closing the door: wouldn't I have violated the law? And think of how often we do things without intending to do them: how we unintentionally infringe copyrights, annoy friends, kill insects. Don't we also sometimes unintentionally close a door? Like, by opening a window? Surely we do.
If the correct applicability of causatives, and thereby of "causing", varies from context to context, should we develop contextualist accounts of causation? I don't think so. Perhaps we should rather do what is common in theories of knowledge and belief: distinguish between real knowledge and belief and our ordinary use of "knowing" and "believing":
"Fred believes that the woman who gave the horrible talk yesterday is a brilliant speaker"; "Fred believes that ophthalmologists are ear doctors"; "Fred believes that I am somebody else". All these can be truly uttered even though Fred does not really believe any of the propositions expressed by the embedded sentences. The thing to do here is to pin down the concept of real belief, and then give an account of belief sentences that draws partly on the subject's real beliefs, but also on various other facts about the subject and the world and the utterance context. (See the first half of the section "Content" in Plurality of Worlds for an account explicitly along these lines.)
So perhaps we should do the same with causation: identify a stable, context-independent, theoretically interesting concept of real causation, and then analyse our ordinary use of "cause" in terms of this relation together with various other facts about the events and the world and the utterance context.
By the way, what would Jaegwon Kim, who feels that real causation is something that pushes and pulls and produces, say about the retina? It seems that the light doesn't produce the activity in the ganglion cells. So are none of our visual perceptions caused by the scenes before us? Or is it context-dependent whether they are so caused?